AI Paper Atlas

Research map / Video / Video segmentation and tracking

Spherical Vision Transformers for Audio-Visual Saliency Prediction in 360$^{\circ }$∘ Videos

Mert Cokelek, Halit Ozsoy, Nevrez İmamoğlu, Cagri Ozcinar, İnci Ayhan, Erkut Erdem, Aykut Erdem

TPAMI 2025 · Video › Video segmentation and tracking · approach: Layered video decomposition and dynamics · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add spherical position embeddings in place of 1D learnable embeddings (ablation)SalViT360 (VSA) · VR-EyeTrackingNSS: 2.518 -> 2.575Improved
add viewport spatio-temporal attention (VSTA) over spatial-only VSA (ablation)SalViT360 · VR-EyeTrackingNSS: 2.575 -> 2.664Improved
add viewport spatial attention over no attention (ablation)SalViT360 · VR-EyeTrackingNSS: 2.306 -> 2.575Improved
add VAC loss with mask on top of VSTA (ablation)SalViT360 · VR-EyeTrackingKLD: 6.174 -> 5.744Improved
add mask to the VAC loss (ablation)SalViT360 · VR-EyeTrackingKLD: 6.011 -> 5.744Improved
add VAC loss without mask on NSS (ablation)SalViT360 · VR-EyeTrackingNSS: 2.664 -> 2.624Hurt
use VSTA instead of 2+1D CNN encoder (ablation)SalViT360 · VR-EyeTrackingNSS: 2.568 -> 2.664Improved
use VSTA instead of offline EMA aggregation (ablation)SalViT360 · VR-EyeTrackingNSS: 2.591 -> 2.664Improved
use VSTA instead of 2+1D CNN on KLD (ablation)SalViT360 · VR-EyeTrackingKLD: 5.915 -> 6.174Hurt
use ambisonics instead of mute ground-truth saliencySalViT360 · YT360-EyeTrackingNSS: 2.115 -> 2.346Improved
add spatial audio branch to visual modelSalViT360 -> SalViT360-AV · YT360-EyeTrackingNSS: 2.346 -> 2.449Improved
use ambisonic encoding instead of raw FOA input (ablation)SalViT360-AV · YT360-EyeTrackingNSS: 2.443 -> 2.449No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy